Legal-RAG
1 What is Legal-RAG?
Legal-RAG is an open-source, end-to-end legal Retrieval-Augmented Generation (RAG) system centered on statutory text. It integrates QueryType-aware routing, hybrid retrieval, bounded graph-augmented context expansion, and provider-agnostic generation. Running OpenAI models with an OPENAI_API_KEY as the generation model is optional; you can deploy with GPU to enable local models (default: Qwen), and other open-source models are configurable.